Traditional diagnosis of Alzheimer Disease (AD) has depended on the postmortem identification of beta-amyloid plaques, tau neurofibrillary tangles, and neuronal loss. Recent efforts to characterize the progression of AD in the aging brain have revealed significant degradation of specific myelinated (white matter) tracts. This pattern of degradation is in close accordance with the spatial distribution of late-myelinating tracts that interconnect associative regions of the cortex. Interestingly, patterns of amyloid deposition in AD appear to follow a similar trajectory. This convergence of observations has led to the hypothesis that myelin degradation may be a primary component of AD, impairing brain connectivity and contributing to deficits in memory, language, and visuospatial processing. These effects tend to be more pronounced and develop earlier in subjects defined as "at risk", based on a family history of the disease and/or carriage of the APOE e4 allele. Current treatment options aimed at slowing disease progression are highly dependent on early diagnosis and drug deployment. Hence, an extensive research program is underway to detect manifestations of AD that appear in the brains of asymptomatic at-risk individuals and to longitudinally track these changes. Central to this effort will be assessments of a) differences in white matter integrity between healthy aging subjects and at-risk subjects using diffusion-weighted MRI and b) alterations in functional connectivity in the same subjects performing an episodic memory task during functional MRI scanning. Integration of structural and functional neuroimaging modalities will allow assessment of impairments in network connectivity. This connectivity profiling should contribute to more reliable early detection of AD and emphasize the importance of myelin-protecting treatment efforts. Relevance to public health: Early detection of Alzheimer Disease is largely dependent on non-invasive neuroimaging techniques that reveal altered structure and function in the brains of subjects with higher risk for developing the disease before overt symptoms have manifested. The integration of new neuroimaging techniques for studying connectivity in the brain is beginning to reveal patterns of degradation affecting specific tissue types. Understanding the trajectory of this degradation should improve early detection of the disease and focus treatment on slowing disease progression in the vulnerable tissues.